Implement lxml parser
This commit is contained in:
Tulir Asokan
@@ -68,6 +68,10 @@ def cut_long_message(message: str, entities: List[TypeMessageEntity]) -> ParsedM
|
||||
return message, entities
|
||||
|
||||
|
||||
class FormatError(Exception):
|
||||
pass
|
||||
|
||||
|
||||
def matrix_to_telegram(html: str) -> ParsedMessage:
|
||||
try:
|
||||
html = command_regex.sub(r"<command>\1</command>", html)
|
||||
@@ -82,8 +86,8 @@ def matrix_to_telegram(html: str) -> ParsedMessage:
|
||||
text, entities = cut_long_message(text, entities)
|
||||
|
||||
return text, entities
|
||||
except Exception:
|
||||
log.exception("Failed to convert Matrix format:\nhtml=%s", html)
|
||||
except Exception as e:
|
||||
raise FormatError(f"Failed to convert Matrix format: {html}") from e
|
||||
|
||||
|
||||
def matrix_reply_to_telegram(content: dict, tg_space: int, room_id: Optional[str] = None
|
||||
|
||||
@@ -14,39 +14,312 @@
|
||||
#
|
||||
# You should have received a copy of the GNU Affero General Public License
|
||||
# along with this program. If not, see <https://www.gnu.org/licenses/>.
|
||||
from lxml import etree
|
||||
from typing import Optional, List, Tuple, Type, Callable, Dict, Any
|
||||
import math
|
||||
import re
|
||||
import logging
|
||||
from typing import Optional, List, Tuple, Union
|
||||
from lxml import html
|
||||
|
||||
from telethon.tl.types import (MessageEntityMention, MessageEntityMentionName, MessageEntityEmail,
|
||||
MessageEntityUrl, MessageEntityTextUrl, MessageEntityBold,
|
||||
MessageEntityItalic, MessageEntityCode, MessageEntityPre,
|
||||
MessageEntityBotCommand, TypeMessageEntity)
|
||||
from telethon.tl.types import (MessageEntityMention as Mention,
|
||||
MessageEntityMentionName as MentionName, MessageEntityEmail as Email,
|
||||
MessageEntityUrl as URL, MessageEntityTextUrl as TextURL,
|
||||
MessageEntityBold as Bold, MessageEntityItalic as Italic,
|
||||
MessageEntityCode as Code, MessageEntityPre as Pre,
|
||||
MessageEntityBotCommand as Command, TypeMessageEntity,
|
||||
InputMessageEntityMentionName as InputMentionName)
|
||||
|
||||
from ...context import Context
|
||||
from ... import user as u, puppet as pu, portal as po
|
||||
from ...db import Message as DBMessage
|
||||
from ...formatter.util import (add_surrogates, remove_surrogates, trim_reply_fallback_html,
|
||||
trim_reply_fallback_text, html_to_unicode)
|
||||
from .parser_common import MatrixParserCommon, ParsedMessage
|
||||
|
||||
|
||||
class MatrixParser(MatrixParserCommon):
|
||||
def __init__(self):
|
||||
self.text = "" # type: str
|
||||
self.entities = [] # type: List[TypeMessageEntity]
|
||||
|
||||
def parse_node(self, node) -> ParsedMessage:
|
||||
pass
|
||||
|
||||
def feed(self, html: str):
|
||||
document = etree.parse(html)
|
||||
self.text, self.entities = self.parse_node(document)
|
||||
|
||||
|
||||
def parse_html(html: str) -> ParsedMessage:
|
||||
parser = MatrixParser()
|
||||
parser.feed(html)
|
||||
return parser.text, parser.entities
|
||||
return MatrixParser.parse(html)
|
||||
|
||||
|
||||
class Entity:
|
||||
@staticmethod
|
||||
def copy(entity: TypeMessageEntity) -> TypeMessageEntity:
|
||||
kwargs = {
|
||||
"offset": entity.offset,
|
||||
"length": entity.length,
|
||||
}
|
||||
if isinstance(entity, Pre):
|
||||
kwargs["language"] = entity.language
|
||||
elif isinstance(entity, TextURL):
|
||||
kwargs["url"] = entity.url
|
||||
elif isinstance(entity, (MentionName, InputMentionName)):
|
||||
kwargs["user_id"] = entity.user_id
|
||||
return entity.__class__(**kwargs)
|
||||
|
||||
@classmethod
|
||||
def adjust(cls, entity: Union[TypeMessageEntity, List[TypeMessageEntity]], *,
|
||||
length: int = None, offset: int = None, offset_diff: int = None
|
||||
) -> Union[TypeMessageEntity, List[TypeMessageEntity]]:
|
||||
if isinstance(entity, list):
|
||||
return [Entity.adjust(element, length=length, offset=offset, offset_diff=offset_diff)
|
||||
for element in entity]
|
||||
entity = cls.copy(entity)
|
||||
if length is not None:
|
||||
entity.length = length
|
||||
if offset is not None:
|
||||
entity.offset = offset
|
||||
if offset_diff is not None:
|
||||
entity.offset += offset_diff
|
||||
if entity.offset < 0:
|
||||
entity.length += entity.offset
|
||||
entity.offset = 0
|
||||
return entity
|
||||
|
||||
|
||||
class TelegramMessage:
|
||||
def __init__(self, text: str = "", entities: Optional[List[TypeMessageEntity]] = None):
|
||||
self.text = text # type: str
|
||||
self.entities = entities or [] # type: List[TypeMessageEntity]
|
||||
|
||||
def offset_entities(self, offset: int) -> "TelegramMessage":
|
||||
def apply_offset(entity: TypeMessageEntity, offset: int):
|
||||
entity = Entity.copy(entity)
|
||||
entity.offset += offset
|
||||
if entity.offset < 0:
|
||||
entity.offset = 0
|
||||
elif entity.offset > len(self.text):
|
||||
return None
|
||||
elif entity.offset + entity.length > len(self.text):
|
||||
entity.length = len(self.text) - entity.offset
|
||||
return entity
|
||||
|
||||
self.entities = [apply_offset(entity, offset) for entity in self.entities if entity]
|
||||
return self
|
||||
|
||||
def append(self, *args: Union[str, "TelegramMessage"]) -> "TelegramMessage":
|
||||
for msg in args:
|
||||
if isinstance(msg, str):
|
||||
msg = TelegramMessage(text=msg)
|
||||
self.entities += Entity.adjust(msg.entities, offset_diff=len(self.text))
|
||||
self.text += msg.text
|
||||
return self
|
||||
|
||||
def prepend(self, *args: Union[str, "TelegramMessage"]) -> "TelegramMessage":
|
||||
for msg in args:
|
||||
if isinstance(msg, str):
|
||||
msg = TelegramMessage(text=msg)
|
||||
self.entities = msg.entities + Entity.adjust(self.entities, offset_diff=len(msg.text))
|
||||
self.text = msg.text + self.text
|
||||
return self
|
||||
|
||||
def format(self, entity_type: type(TypeMessageEntity), offset: int = None, length: int = None,
|
||||
**kwargs) -> "TelegramMessage":
|
||||
self.entities.append(entity_type(offset=offset or 0,
|
||||
length=length if length is not None else len(self.text),
|
||||
**kwargs))
|
||||
return self
|
||||
|
||||
def concat(self, *args: Union[str, "TelegramMessage"]) -> "TelegramMessage":
|
||||
return TelegramMessage().append(self, *args)
|
||||
|
||||
def trim(self) -> "TelegramMessage":
|
||||
orig_len = len(self.text)
|
||||
self.text = self.text.lstrip()
|
||||
diff = orig_len - len(self.text)
|
||||
self.text = self.text.rstrip()
|
||||
self.offset_entities(-diff)
|
||||
return self
|
||||
|
||||
def split(self, separator, max_items: int = 0) -> List["TelegramMessage"]:
|
||||
text_parts = self.text.split(separator, max_items - 1)
|
||||
output = [] # type: List[TelegramMessage]
|
||||
|
||||
offset = 0
|
||||
for part in text_parts:
|
||||
msg = TelegramMessage(part)
|
||||
for entity in self.entities:
|
||||
start_in_range = len(part) > entity.offset - offset >= 0
|
||||
end_in_range = len(part) >= entity.offset - offset + entity.length > 0
|
||||
if start_in_range and end_in_range:
|
||||
msg.entities.append(Entity.adjust(entity, offset_diff=-offset))
|
||||
output.append(msg)
|
||||
|
||||
offset += len(part)
|
||||
offset += len(separator)
|
||||
|
||||
return output
|
||||
|
||||
@staticmethod
|
||||
def join(items: List[Union[str, "TelegramMessage"]], separator: str = " ") -> "TelegramMessage":
|
||||
main = TelegramMessage()
|
||||
for msg in items:
|
||||
if isinstance(msg, str):
|
||||
msg = TelegramMessage(text=msg)
|
||||
main.entities += Entity.adjust(msg.entities, offset_diff=len(main.text))
|
||||
main.text += msg.text + separator
|
||||
main.text = main.text[:-len(separator)]
|
||||
return main
|
||||
|
||||
|
||||
class MatrixParser(MatrixParserCommon):
|
||||
@classmethod
|
||||
def list_to_tmessage(cls, node: html.HtmlElement, strip_linebreaks) -> TelegramMessage:
|
||||
ordered = node.tag == "ol"
|
||||
tagged_children = cls.node_to_tagged_tmessages(node, strip_linebreaks)
|
||||
counter = 1
|
||||
indent_length = 0
|
||||
if ordered:
|
||||
try:
|
||||
counter = int(node.attrib.get("start", "1"))
|
||||
except ValueError:
|
||||
counter = 1
|
||||
|
||||
longest_index = counter - 1 + len(tagged_children)
|
||||
indent_length = len(str(longest_index))
|
||||
indent = (indent_length + 4) * " "
|
||||
children = [] # type: List[TelegramMessage]
|
||||
for child, tag in tagged_children:
|
||||
if tag != "li":
|
||||
continue
|
||||
|
||||
if ordered:
|
||||
prefix = f"{counter}. "
|
||||
counter += 1
|
||||
else:
|
||||
prefix = "● "
|
||||
child = child.prepend(prefix)
|
||||
parts = child.split("\n")
|
||||
parts = parts[:1] + [part.prepend(indent) for part in parts[1:]]
|
||||
child = TelegramMessage.join(parts, "\n")
|
||||
children.append(child)
|
||||
return TelegramMessage.join(children, "\n")
|
||||
|
||||
@classmethod
|
||||
def blockquote_to_tmessage(cls, node: html.HtmlElement, strip_linebreaks) -> TelegramMessage:
|
||||
msg = cls.tag_aware_parse_node(node, strip_linebreaks)
|
||||
children = msg.trim().split("\n")
|
||||
children = [child.prepend("> ") for child in children]
|
||||
return TelegramMessage.join(children, "\n")
|
||||
|
||||
@classmethod
|
||||
def header_to_tmessage(cls, node: html.HtmlElement, strip_linebreaks) -> TelegramMessage:
|
||||
children = cls.node_to_tmessages(node, strip_linebreaks)
|
||||
length = int(node.tag[1])
|
||||
prefix = "#" * length + " "
|
||||
return TelegramMessage.join(children, "").prepend(prefix)
|
||||
|
||||
@classmethod
|
||||
def basic_format_to_tmessage(cls, node: html.HtmlElement, strip_linebreaks) -> TelegramMessage:
|
||||
msg = cls.tag_aware_parse_node(node, strip_linebreaks)
|
||||
if node.tag in ("b", "strong"):
|
||||
msg.format(Bold)
|
||||
elif node.tag in ("i", "em"):
|
||||
msg.format(Italic)
|
||||
elif node.tag == "command":
|
||||
msg.format(Command)
|
||||
elif node.tag in ("s", "del"):
|
||||
pass # TODO
|
||||
elif node.tag in ("u", "ins"):
|
||||
pass # TODO
|
||||
return msg
|
||||
|
||||
@classmethod
|
||||
def link_to_tstring(cls, node: html.HtmlElement, strip_linebreaks) -> TelegramMessage:
|
||||
msg = cls.tag_aware_parse_node(node, strip_linebreaks)
|
||||
href = node.attrib.get("href", "")
|
||||
if not href:
|
||||
return msg
|
||||
|
||||
if href.startswith("mailto:"):
|
||||
return TelegramMessage(href[len("mailto:"):]).format(Email)
|
||||
|
||||
mention = cls.mention_regex.match(href)
|
||||
if mention:
|
||||
mxid = mention.group(1)
|
||||
user = (pu.Puppet.get_by_mxid(mxid)
|
||||
or u.User.get_by_mxid(mxid, create=False))
|
||||
if not user:
|
||||
return msg
|
||||
if user.username:
|
||||
return TelegramMessage(f"@{user.username}").format(Mention)
|
||||
elif user.tgid:
|
||||
return TelegramMessage(user.displayname or msg.text).format(MentionName,
|
||||
user_id=user.tgid)
|
||||
return msg
|
||||
|
||||
room = cls.room_regex.match(href)
|
||||
if room:
|
||||
username = po.Portal.get_username_from_mx_alias(room.group(1))
|
||||
portal = po.Portal.find_by_username(username)
|
||||
if portal and portal.username:
|
||||
return TelegramMessage(f"@{portal.username}").format(Mention)
|
||||
|
||||
return (msg.format(URL)
|
||||
if msg.text == href
|
||||
else msg.format(TextURL, url=href))
|
||||
|
||||
@classmethod
|
||||
def node_to_tmessage(cls, node: html.HtmlElement, strip_linebreaks) -> TelegramMessage:
|
||||
if node.tag == "blockquote":
|
||||
return cls.blockquote_to_tmessage(node, strip_linebreaks)
|
||||
elif node.tag in ("ol", "ul"):
|
||||
return cls.list_to_tmessage(node, strip_linebreaks)
|
||||
elif node.tag in ("h1", "h2", "h3", "h4", "h5", "h6"):
|
||||
return cls.header_to_tmessage(node, strip_linebreaks)
|
||||
elif node.tag == "br":
|
||||
return TelegramMessage("\n")
|
||||
elif node.tag in ("b", "strong", "i", "em", "s", "del", "u", "ins", "command"):
|
||||
return cls.basic_format_to_tmessage(node, strip_linebreaks)
|
||||
elif node.tag == "a":
|
||||
return cls.link_to_tstring(node, strip_linebreaks)
|
||||
elif node.tag == "p":
|
||||
return cls.tag_aware_parse_node(node, strip_linebreaks).append("\n")
|
||||
elif node.tag == "pre":
|
||||
lang = ""
|
||||
try:
|
||||
if node[0].tag == "code":
|
||||
lang = node[0].attrib["class"][len("language-"):]
|
||||
node = node[0]
|
||||
except (IndexError, KeyError):
|
||||
pass
|
||||
return cls.parse_node(node, strip_linebreaks=False).format(Pre, language=lang)
|
||||
elif node.tag == "code":
|
||||
return cls.parse_node(node, strip_linebreaks=False).format(Code)
|
||||
return cls.tag_aware_parse_node(node, strip_linebreaks)
|
||||
|
||||
@staticmethod
|
||||
def text_to_tmessage(text: str, strip_linebreaks: bool = True) -> TelegramMessage:
|
||||
if strip_linebreaks:
|
||||
text = text.replace("\n", "")
|
||||
return TelegramMessage(text)
|
||||
|
||||
@classmethod
|
||||
def node_to_tagged_tmessages(cls, node: html.HtmlElement, strip_linebreaks: bool = True
|
||||
) -> List[Tuple[TelegramMessage, str]]:
|
||||
output = []
|
||||
|
||||
if node.text:
|
||||
output.append((cls.text_to_tmessage(node.text, strip_linebreaks), "text"))
|
||||
for child in node:
|
||||
output.append((cls.node_to_tmessage(child, strip_linebreaks), child.tag))
|
||||
if child.tail:
|
||||
output.append((cls.text_to_tmessage(child.tail, strip_linebreaks), "text"))
|
||||
return output
|
||||
|
||||
@classmethod
|
||||
def node_to_tmessages(cls, node: html.HtmlElement, strip_linebreaks) -> List[
|
||||
TelegramMessage]:
|
||||
return [msg for (msg, tag) in cls.node_to_tagged_tmessages(node, strip_linebreaks)]
|
||||
|
||||
@classmethod
|
||||
def tag_aware_parse_node(cls, node: html.HtmlElement,
|
||||
strip_linebreaks: bool = True) -> TelegramMessage:
|
||||
msgs = cls.node_to_tagged_tmessages(node, strip_linebreaks)
|
||||
output = TelegramMessage()
|
||||
for msg, tag in msgs:
|
||||
if tag in cls.block_tags:
|
||||
msg = msg.append("\n").prepend("\n")
|
||||
output = output.append(msg)
|
||||
return output.trim()
|
||||
|
||||
@classmethod
|
||||
def parse_node(cls, node: html.HtmlElement, strip_linebreaks: bool = True) -> TelegramMessage:
|
||||
return TelegramMessage.join(cls.node_to_tmessages(node, strip_linebreaks))
|
||||
|
||||
@classmethod
|
||||
def parse(cls, data: str) -> ParsedMessage:
|
||||
document = html.fromstring(f"<html>{data}</html>")
|
||||
msg = cls.parse_node(document)
|
||||
return msg.text, msg.entities
|
||||
|
||||
Reference in New Issue
Block a user